PKU High-Dimensional Data Visualization - Galleries

Dimension Reconstruction for Visual Exploration of Subspace Clusters in High-dimensional Data

In the projection, users can define a new dimension (like RD1 and RD2), where subspace clusters can be separated. Then, he can join the new dimensions in an original subspace (Subspace 3) to maintain the separation of data clusters.

Dimension Projection-Matrix/Tree: Interactive Subspace Visual Exploration and Analysis of High Dimensional Data

In the dimension projection matrix (1), each cell is either a data projection (top right corner) or a dimension projection (bottom left corner) of a subspace. Rows and columns represent dimensions that constitute the subspaces.

The dimension projection tree: users can brush a group of data items (left) or dimensions (right) to create a child node containing the corresponding subspace.

Visualization Assembly Line

MLMD: Multi-Layered Visualization for Multi-Dimensional Data

Multi-Dimensional Transfer Function Design based on Flexible Dimension Projection Embedded in Parallel Coordinates

Multi-Dimensional Transfer Function Design based on Flexible Dimension Projection Embedded in Parallel Coordinates

Interactive Local Clustering Operations In Parallel Coordinates

Scattering Points in Parallel Coordinates

Splatting the Lines in Parallel Coordinates

Visual Clustering in Parallel Coordinates

---